ASCO 2016

Targeting PD-L1 with avelumab in metastatic Merkel cell carcinoma

Dr Howard Kaufman - Rutgers Cancer Institute of New Jersey, USA

I led the phase II study of avelumab in the treatment of advanced Merkel cell carcinoma.

Could you tell us more about these trials and the results?

So Merkel cell carcinoma is a rare form of skin cancer but it’s one of the most deadly types of cancer so most patients if they develop metastatic disease usually succumb to the disease within 6-12 months. There’s no real standard of care for treating this disease although the tumour is sensitive to chemotherapy and radiation but initial response rates with those approaches usually are not associated with long-term survival. So for a while we’ve known that Merkel cell might be amenable to immunotherapy. This has been based on the fact that the tumour often arises in immunosuppressed individuals, sometimes the primary will regress with the immune system and then will become metastatic at a later date. More recently we’ve identified PD-L1 expression within the tumour cells and we now know that PD-L1 expression often is associated with better outcomes with immunotherapy, particularly with the T-cell checkpoint inhibitors.

Avelumab is a PD-L1 monoclonal antibody. It’s based on an IgG1 isotype so in addition to blocking the PD-1 PD-L1 interaction it also may have ADCC activity, so it may bring in NK cells as well. So in this study patients who had already failed at least one line of prior chemotherapy and had metastatic Merkel cell carcinoma were enrolled in treatment. We consented 125 patients and eventually treated 88 of these patients on trial. The medication is given as a 30 minute IV infusion every two weeks and we kept patients on the treatment until they either had a confirmed objective response if they had progressive disease or if there were side effects. What we found in the study is that we had an objective response rate of approximately 32% and an additional 10% of patients had stable disease at the end of the study. At the time of the analysis, and our median follow up right now is at 10 months, 80% of the responders were still in response and this suggests that there is going to be durable long term responses associated with this drug as we’ve seen with other immunotherapy agents.

In terms of the side effects, what we saw was generally low grade toxicity. The most frequent things we saw were fatigue and infusion type reactions which could be prevented with pre-medication. There were a small number of events of immune mediated adverse events that had been reported with other T-cell checkpoint inhibitors but these were generally low grade. There was only one grade 3 or greater event in the study and that was an isolated lab abnormality of high hepatic transaminases.

So this treatment has the opportunity to really change the standard of care for these patients. We saw no impact of either PD-L1 expression or the presence of the Merkel cell polyomavirus which is often associated with this tumour in terms of getting a response. So it seemed to work in every patient, regardless of these other variables. We did see a slightly higher response rate in patients who had only one prior line of therapy, so there was a 40% response rate in that group compared to about 19% in patients who had two or more lines of therapy. So one of the important directions in the future will be to maybe bring this drug to earlier stage disease. I think there’s a big opportunity to take this into the adjuvant setting and then we’re quite excited about doing combination studies with this agent and other immunotherapy regimens.

I spoke with Dr Paul Nghiem at AACR this year, I think he was working with nivolumab and Merkel cell carcinoma. Do you find these results to be complementary, maybe even combinable?

Yes, there’s a recent study in which the pembrolizumab was used in first line Merkel cell and although the number of patients in that study was smaller than in our study there were nonetheless very exciting response rates in the range of 50-55%. So it really suggests that the PD-1 PD-L1 axis is going to be an appropriate target for patients and so I don’t think we can have enough good drugs available for these patients. We don’t really understand if patients who respond to or don’t respond to one of these could respond to the other one so I think both results are finding and both results are consistent. I think it’s an exciting time for patients with cancer, given the results that we’re seeing.

I think our study in Merkel cell has one other important implication. Merkel cell is often associated with a virus which is the Merkel cell polyomavirus. Our results might suggest that other tumours that have a viral association might be amenable to this type of immunotherapy. The reason this might be particularly interesting is that these are the tumours that often do not have a high mutation burden with the neoantigens that have been previously suggested as important in getting response to the checkpoint inhibitors. So the mechanism of getting this response may be a little bit different but it obviously opens up the door to other cancers where viruses are associated, such as the HPV and head and neck cancer and cervical cancer, EBV with nasopharyngeal carcinoma and some gastric cancers. In fact there’s some early data from phase I studies with avelumab that these populations might in fact be responsive to treatment.